CN111063083A

CN111063083A - Access control method and device, computer readable storage medium and computer equipment

Info

Publication number: CN111063083A
Application number: CN201911295842.3A
Authority: CN
Inventors: 泮诚
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-12-16
Filing date: 2019-12-16
Publication date: 2020-04-24
Anticipated expiration: 2039-12-16
Also published as: CN111063083B

Abstract

The application relates to deep learning, and provides an access control method, an access control device, a computer-readable storage medium and computer equipment, wherein the method comprises the following steps: acquiring a monitoring video frame sequence, and determining a face image to be processed corresponding to the face identifier to be processed and information of each body frame to be processed according to the monitoring video frame sequence; acquiring time intervals of monitoring video frames, and determining human body advancing speed information according to the information of each human body frame to be processed and the time intervals; determining human body advancing direction information according to the information of each human body frame to be processed, and obtaining a door opening intention result when the human body advancing direction information and the human body advancing speed information accord with preset conditions; when a door opening intention result is obtained, carrying out face recognition on a face image to be processed to obtain a face recognition passing result; and obtaining a door opening instruction according to the face recognition passing result and the door opening intention result, and executing door opening operation by the access control equipment according to the door opening instruction. The access control device and the access control method improve the accuracy of control over the access control device.

Description

Access control method and device, computer readable storage medium and computer equipment

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for controlling an access control, a computer-readable storage medium, and a computer device.

Background

With the development of face recognition technology, face recognition is applied to various aspects of people's life. For example, the access control device can be controlled by a face recognition technology, so that intelligent access control is realized. The existing method for controlling the access control by face recognition generally comprises the steps of recognizing faces and judging the identities of people corresponding to the faces, and controlling an access control device to intelligently open a door when identity verification is passed.

However, in the existing intelligent door opening method, the situation of door opening by mistake still exists, so that the door access control cannot be accurately carried out.

Disclosure of Invention

Accordingly, it is necessary to provide an access control method, an apparatus, a computer-readable storage medium, and a computer device for solving the technical problem that access control cannot be performed accurately.

An access control method, comprising:

acquiring a monitoring video frame sequence, and determining a face image to be processed corresponding to the face identifier to be processed and information of each body frame to be processed according to the monitoring video frame sequence;

acquiring time intervals of monitoring video frames, and determining human body advancing speed information corresponding to human body identification to be processed according to the human body frame information to be processed and the time intervals;

determining human body advancing direction information corresponding to the human body identifier to be processed according to the information of each human body frame to be processed, and obtaining a door opening intention result corresponding to the human body identifier to be processed when the human body advancing direction information and the human body advancing speed information meet preset conditions;

when a door opening intention result is obtained, matching the face image to be processed with a preset face library, and when the matching is successful, obtaining a face recognition passing result;

and obtaining a door opening instruction according to the face recognition passing result and the door opening intention result, and sending the door opening instruction to the access control equipment so that the access control equipment executes door opening operation according to the door opening instruction.

An access control device comprising:

the information determining module is used for acquiring a monitoring video frame sequence and determining a to-be-processed face image corresponding to the to-be-processed face identification and information of each to-be-processed human body frame according to the monitoring video frame sequence;

the speed determining module is used for acquiring the time interval of the monitoring video frame and determining the human body advancing speed information corresponding to the human body identifier to be processed according to the information of each human body frame to be processed and the time interval;

the intention obtaining module is used for determining human body advancing direction information corresponding to the human body identification to be processed according to the information of each human body frame to be processed, and obtaining a door opening intention result corresponding to the human body identification to be processed when the human body advancing direction information and the human body advancing speed information meet preset conditions;

the face recognition module is used for matching the face image to be processed with a preset face library when the door opening intention result is obtained, and obtaining a face recognition passing result when the matching is successful;

and the access control module is used for obtaining a door opening instruction according to the face recognition passing result and the door opening intention result and sending the door opening instruction to the access control equipment so that the access control equipment executes door opening operation according to the door opening instruction.

A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the program:

A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, causes the processor to perform the steps of:

According to the access control method, the access control device, the computer readable storage medium and the computer equipment, the to-be-processed face image corresponding to the to-be-processed face identification and the information of each to-be-processed human body frame are determined according to the monitoring video frame sequence by acquiring the monitoring video frame sequence; determining human body advancing speed information according to the information of each human body frame to be processed and the time interval by acquiring the time interval of the monitoring video frame; determining human body advancing direction information according to the information of each human body frame to be processed, and obtaining a door opening intention result corresponding to the human body identifier to be processed when the human body advancing direction information and the human body advancing speed information meet preset conditions; when a door opening intention result is obtained, matching the face image to be processed with a preset face library, and when the matching is successful, obtaining a face recognition passing result; and obtaining a door opening instruction according to the face recognition passing result and the door opening intention result, and sending the door opening instruction to the access control equipment so that the access control equipment executes door opening operation according to the door opening instruction. Whether the door opening intention exists is determined by determining the advancing speed and the advancing direction of the human body, the face recognition result is obtained through face recognition, the access control equipment is controlled according to the door opening intention and the face recognition result, the mistaken door opening operation of the access control equipment can be reduced, and the accuracy of controlling the access control equipment is improved.

Drawings

Fig. 1 is an application environment diagram of an access control method in one embodiment;

FIG. 2 is a schematic flow chart illustrating a method for controlling access control according to an embodiment;

FIG. 2A is a schematic flow chart illustrating the determination of a to-be-processed face image corresponding to a to-be-processed face identifier and information of each to-be-processed body frame according to an embodiment;

FIG. 3 is a schematic flow chart illustrating the determination of the area of a target face frame according to an embodiment;

FIG. 4 is a schematic flow chart illustrating the process of obtaining human body travel speed information according to one embodiment;

FIG. 5 is a schematic illustration of a human body undergoing a forward deceleration motion in one embodiment;

FIG. 6 is a schematic illustration of a human body undergoing forward acceleration in one embodiment;

FIG. 7 is a schematic diagram of a human body performing a forward uniform motion in an exemplary embodiment;

FIG. 8 is a timing diagram illustrating a method for controlling access control in an exemplary embodiment;

FIG. 9 is a schematic flow chart of a method for controlling access in an embodiment;

FIG. 10 is a schematic view of the embodiment of FIG. 9 showing the door opening success;

FIG. 11 is a schematic view of the embodiment of FIG. 9 showing a door open failure;

fig. 12 is a block diagram showing the construction of an access control device according to an embodiment;

FIG. 13 is a block diagram of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

Computer Vision technology (CV) Computer Vision is a science for researching how to make a machine "see", and further refers to that a camera and a Computer are used to replace human eyes to perform machine Vision such as identification, tracking and measurement on a target, and further image processing is performed, so that the Computer processing becomes an image more suitable for human eyes to observe or transmitted to an instrument to detect. As a scientific discipline, computer vision research-related theories and techniques attempt to build artificial intelligence systems that can capture information from images or multidimensional data. Computer vision technologies generally include image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content/behavior recognition, three-dimensional object reconstruction, 3D technologies, virtual reality, augmented reality, synchronous positioning, map construction, and other technologies, and also include common biometric technologies such as face recognition and fingerprint recognition.

Machine Learning (ML) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and formal education learning.

The scheme provided by the embodiment of the application relates to the computer vision technology, the machine learning technology and other technologies of artificial intelligence, and is specifically explained by the following embodiments:

fig. 1 is an application environment diagram of the access control method in one embodiment. Referring to fig. 1, the access control method is applied to an access control system. The access control system includes a monitoring device 102, a server 104, and an access control device 106. The monitoring device 102 and the server 104 are connected through a network, and the access control device 106 and the server 104 are connected through a network. The monitoring device may specifically be a camera, a video camera, etc. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers. The entrance guard equipment is used for controlling entrance guard and opening the door.

Specifically, the server 104 obtains a monitoring video stream sent by the monitoring device 102, obtains a monitoring video frame sequence according to the monitoring video stream, and determines the to-be-processed face image and the information of each to-be-processed body frame corresponding to the to-be-processed face identifier according to the monitoring video frame sequence. The server 104 acquires the time interval of the monitoring video frame, and determines the human body advancing speed information corresponding to the human body identifier to be processed according to the information of each human body frame to be processed and the time interval. The server 104 determines human body advancing direction information corresponding to the human body identifier to be processed according to the information of each human body frame to be processed, and obtains a door opening intention result corresponding to the human body identifier to be processed when the human body advancing direction information and the human body advancing speed information meet preset conditions; and when the result of the intention of opening the door is obtained, matching the face image to be processed with a preset face library, and when the matching is successful, obtaining a face recognition passing result. The server 104 obtains a door opening instruction according to the face recognition pass result and the door opening intention result, and sends the door opening instruction to the door control device 106, so that the door control device 106 executes a door opening operation according to the door opening instruction.

In one embodiment, the access control method is applied to an access control system, where the access control system includes a monitoring device capable of performing edge calculation and an access control device, where the device capable of performing edge calculation may specifically be an AI camera, TX1 (edge AI module), TK1 (edge AI module), Movidias (edge AI module), and the like.

As shown in fig. 2, in one embodiment, an access control method is provided. The embodiment is mainly illustrated by applying the method to the server 104 in fig. 1. Referring to fig. 2, the access control method specifically includes the following steps:

s202, acquiring a monitoring video frame sequence, and determining a to-be-processed face image corresponding to the to-be-processed face identification and information of each to-be-processed human body frame according to the monitoring video frame sequence.

The monitoring video frame sequence refers to each monitoring video frame obtained by converting a monitoring video stream acquired by monitoring equipment. The face identifier to be processed is used for uniquely identifying the face to be processed in the monitoring video frame. For example, the face identifier corresponding to the maximum face frame area may be obtained as the face identifier to be processed according to the face frame areas in the monitored video frame sequence, or the face identifiers in the monitored video frame sequence may be sorted according to the face frame areas in the monitored video frame sequence to obtain the face identifier sequence, and the face identifier to be processed is sequentially selected from the face identifiers. The face image to be processed refers to a face image corresponding to the face identifier to be processed. The body frame information is information related to the obtained body frame. Such as the position of the body frame and the area of the body frame. There may be body frame information in each video frame. The to-be-processed human body frame information refers to human body frame information corresponding to the to-be-processed human body identifier.

Specifically, the monitoring device sends a monitoring video stream to the server, the server receives the monitoring video stream and converts the monitoring video stream into a monitoring video frame sequence, face detection is performed on each monitoring video frame in the monitoring video frame sequence, a to-be-processed face image corresponding to the to-be-processed face identifier is determined, human body detection is performed on each monitoring video frame in the monitoring video frame sequence, and information of each to-be-processed human body frame corresponding to the to-be-processed human body identifier is determined. The human body detection means detecting and positioning a human body in a video frame based on a deep learning algorithm, and returning human body identification and high-precision human body frame information, wherein the deep learning algorithm can be a convolutional neural network, a recurrent neural network, a long-short term memory network and the like. The human body identification is used for uniquely identifying the human body, and the human face identification corresponds to the associated human body identification one to one. The human body identification to be processed refers to the human body identification corresponding to the human face identification to be processed. The to-be-processed human body frame information refers to human body frame information corresponding to the to-be-processed human body identifier.

And S204, acquiring the time interval of the monitoring video frame, and determining the human body advancing speed information corresponding to the human body identifier to be processed according to the information of each human body frame to be processed and the time interval.

The time interval refers to a time interval between every two video frames in the monitoring video sequence, and the time interval may be determined according to the number of frames transmitted per second of the pictures of the video stream, for example, if the transmission speed of the video stream is 25FPS (frames transmitted per second of the pictures), the time interval between every two frames in the video sequence corresponding to the video stream is 40ms (milliseconds). The human body travel speed information refers to travel speed information of a person corresponding to the human body identifier, and may include a human body travel acceleration and a human body travel speed. The human body travel speed may also be the speed between every two video frames in the video sequence.

Specifically, the server determines time intervals among monitoring video frames in the video sequence according to the speed of the monitoring video stream, and determines human body advancing speed information corresponding to the human body identifier to be processed according to the human body positions and the time intervals in the human body frame information to be processed. For example, the distance between the human body positions in two continuous monitoring video frames is determined, and the human body advancing speed between the two continuous monitoring video frames is obtained according to the distance and the time interval.

S206, determining the human body advancing direction information corresponding to the human body identifier to be processed according to the human body frame information to be processed, and obtaining the door opening intention result corresponding to the human body identifier to be processed when the human body advancing direction information and the human body advancing speed information meet the preset conditions.

The human body moving direction information refers to a direction in which the human body moves relative to the monitoring device, for example, forward moving toward the monitoring device is forward, and backward moving away from the monitoring device is reverse. The preset condition refers to a preset condition that the preset detection is the door opening intention. For example, the human body may be considered to have an intention to open the door if the human body is moving forward, the human body is moving negatively and the human body moving speed does not exceed a preset threshold. It is also possible that the human body traveling speed is a specific value, such as 0, i.e. the human body in the sequence of monitored video frames is stationary, i.e. the human body is considered to have an intention to open the door. The result of the intention to open the door means that the human body has an intention to open the door.

Specifically, the server may determine the human body traveling direction information corresponding to the human body identifier to be processed according to the human body frame area in each piece of human body frame information to be processed. In one embodiment, the area of the human body frame in each piece of the human body frame information to be processed is smaller and smaller according to the sequence of the monitoring video frames, namely, the human body travels in the reverse direction. In one embodiment, the area of the human body frame in each piece of to-be-processed human body frame information is larger and larger according to the sequence of the monitoring video frames, and then the human body is in forward direction. When the human body advancing direction information and the human body advancing speed information accord with preset conditions, the server obtains a door opening intention result corresponding to the human body identifier to be processed

And S208, when the door opening intention result is obtained, matching the face image to be processed with a preset face library, and when the matching is successful, obtaining a face recognition passing result.

The preset face library is a preset face database and is used for face recognition, and further determining identity authority, namely determining whether the door is opened or not.

Specifically, when the server obtains the result of the intention to open the door, the server matches the face image to be processed with a preset face library, and when the matching is successful, a face recognition passing result is obtained. That is, the server performs face recognition matching only when there is an intention to open the door, and does not need face recognition matching when there is no intention to open the door. The workload of the server can be reduced, and the efficiency is improved.

And S210, obtaining a door opening instruction according to the face recognition passing result and the door opening intention result, and sending the door opening instruction to the access control equipment so that the access control equipment executes door opening operation according to the door opening instruction.

Wherein, entrance guard's equipment is used for the control door to carry out the switch operation.

Specifically, the server obtains a door opening instruction according to the face recognition passing result and the door opening intention result, sends the door opening instruction to the access control device, and the access control device receives the door opening instruction and executes door opening operation. In one embodiment, the server may send the door opening command to the door access device through a GPIO (General-purpose input/output) signal.

According to the access control equipment control method, whether door opening intention exists or not is determined by determining the advancing speed and the advancing direction of a human body, the face recognition result is obtained through face recognition, the access control equipment is controlled according to whether the door opening intention exists or not and the face recognition result, namely, when the door opening intention result exists and the face recognition passing result is obtained, the door opening instruction is sent to the access control equipment, the mistaken door opening operation of the access control equipment can be reduced, and the accuracy of controlling the access control equipment is improved.

In one embodiment, as shown in fig. 2A, the step S202 of determining the to-be-processed face image and the to-be-processed body frame information corresponding to the to-be-processed face identifier according to the sequence of monitored video frames includes the steps of:

s202a, acquiring a monitoring video frame sequence, performing face detection on the monitoring video frame sequence to obtain each face identifier and each face image, acquiring a face identifier to be processed from each face identifier, and determining a face image to be processed corresponding to the face identifier to be processed from each face image.

S202b, performing human body detection on the monitoring video frame sequence to obtain each human body identifier and each human body frame information, determining the human body identifier to be processed from each human body identifier according to the human face identifier to be processed, and determining each human body frame information to be processed corresponding to the human body identifier to be processed from each human body frame information.

The face detection means that for any given video frame, searching is performed based on a deep learning algorithm to determine whether the video frame contains a face, if so, relevant information of the face is returned, and the relevant information of the face can include a face image, a face frame position, a face frame area, face identification and the like. There may be multiple different faces in a video frame. The face identification is automatically generated and used for uniquely identifying the face in the monitoring video frame, and a face image corresponding to the face identification can be arranged in each video frame.

Specifically, the server performs face detection on the sequence of monitoring video frames to obtain a face identifier and a face image corresponding to each monitoring video frame, obtains a face identifier to be processed from each face identifier, and determines a face image to be processed corresponding to the face identifier to be processed from each face image. And then, performing human body detection on the monitoring video frame sequence to obtain human body identification and human body frame information corresponding to each monitoring video frame, determining the human body identification to be processed from each human body identification according to the human face identification to be processed, and determining each human body frame information to be processed corresponding to the human body identification to be processed from each human body frame information. The accuracy of obtaining the face image to be processed and the information of each body frame to be processed is improved through human body detection and human face recognition.

In one embodiment, as shown in fig. 3, step S202a, namely, performing face detection on the sequence of monitored video frames to obtain each face identifier and each face image, includes the steps of:

s302, inputting the monitoring video frame sequence into the trained face detection model for face detection to obtain each face identification, each face frame image and each corresponding face frame area.

The trained face detection model is a face detection model obtained by training according to existing face data by using a deep learning algorithm. The deep learning algorithm can be Convolutional Neural Network (CNN) -based, cascaded convolutional neural network-based, multitask convolutional neural network-based, and the like. The face frame area is used for representing the size of a face and can be calculated according to the coordinates of the face frame obtained by face detection. For example, the length and width of the face frame are calculated from the coordinates of each point in the face frame, and then the area of the face frame is calculated.

Specifically, the server inputs the sequence of the monitoring video frames into a trained face detection model for face detection, and obtains each face identifier, each face frame image and each corresponding face frame area.

S304, determining the area of each face frame to be processed corresponding to the face image to be processed from the area of each face frame, and determining the area of a target face frame from the area of each face frame to be processed.

The area of the face frame to be processed refers to the area of each face frame in the sequence of the monitoring video frames corresponding to the face identifier to be processed. The target face frame area refers to the largest face frame area to be processed in the face frame areas to be processed.

Specifically, the server determines each area of the face frame to be processed corresponding to the face image to be processed from each area of the face frame, and determines the area of the target face frame from each area of the face frame to be processed.

Step S210, namely, matching the face image to be processed with a preset face library, and obtaining a face recognition passing result when matching is successful, including the steps of:

when the area of the target face frame meets a preset area condition, inputting a face image to be processed into a trained face recognition model for face feature recognition to obtain a face feature recognition result, matching the face feature recognition result with face features in a preset face library, and when the matching is successful, obtaining a face recognition passing result, wherein the condition that the area of the target face frame meets the preset area condition comprises at least one of the condition that the area of the target face frame exceeds a preset area threshold value and the area of the target face frame is equal to the preset target area.

The preset area condition is a preset face frame area capable of carrying out face recognition, and the target face frame area meets the preset area condition and comprises at least one of the target face frame area exceeding a preset area threshold value and the target face frame area being equal to the preset target area. The preset area threshold refers to a preset human face area threshold. The target area refers to a preset minimum face area. The face feature recognition result refers to the face feature recognition of the face image to be processed to obtain the face feature corresponding to the face image to be processed. The face recognition passing result means that the face of the face image to be processed passes through the face recognition. The face recognition passing result comprises a face identity corresponding to the face image to be processed.

Specifically, the server judges whether the area of the target face frame meets a preset area condition, and when the area of the target face frame does not meet the preset area condition, the server considers that the face image to be processed corresponding to the area of the target face frame cannot be subjected to face recognition, and then a face recognition failure result is obtained. When the area of the target face frame accords with a preset area condition, a face image to be processed is input into a trained face recognition model for face feature recognition to obtain a face feature recognition result, the face feature recognition result is matched with face features in a preset face library, and when a face consistent with the face features in the face feature recognition result is matched, a face recognition passing result is obtained.

In one embodiment, when a face consistent with the face features in the face feature recognition result is not matched, a face recognition failure result is obtained. At the moment, the server returns information without permission to enter to the entrance guard display device.

In an embodiment, the face feature recognition result may also be matched with face features in a preset face library to obtain a matching degree between the face feature recognition result and each face in the preset face library, and when the maximum matching degree exceeds a preset threshold, a face identity corresponding to the maximum matching degree is obtained to obtain a face recognition passing result. And when the maximum matching degree does not exceed the preset threshold value, obtaining a face recognition failure result.

In one embodiment, when the area of the target face frame is within a preset threshold range, the face image to be processed is input into a trained face recognition model for face feature recognition to obtain a face feature recognition result, the face feature recognition result is matched with face features in a preset face library, and when the matching is successful, a face recognition passing result is obtained.

In the above embodiment, the area of each face frame is obtained through face detection, the area of the target face frame is determined from the area of each face frame, and when the area of the target face frame exceeds a preset threshold, face recognition is performed, that is, by recognizing a face image to be processed corresponding to the area of the target face frame exceeding the preset threshold, the situation of inaccurate recognition can be reduced, and the accuracy of face recognition is improved.

In one embodiment, after step S304, that is, after determining each face frame area to be processed corresponding to the face image to be processed from each face frame area, the method further includes the steps of:

and determining the face advancing direction information corresponding to the face identification to be processed according to the area of each face frame to be processed.

The face moving direction refers to a direction in which the face moves relative to the monitoring device, for example, forward moving toward the monitoring device is forward, and backward moving away from the monitoring device is reverse.

Specifically, the face traveling direction information corresponding to the face identifier to be processed may be determined according to the size of each face frame area to be processed. For example, when the area of each face frame to be processed is larger and larger according to the sequence of the monitored video frames, the face traveling direction is a forward direction. When the area of each face frame to be processed is smaller and smaller according to the sequence of the monitoring video frames, the advancing direction of the face is reverse.

After step S208, that is, after determining the human body traveling direction information corresponding to the human body identifier to be processed according to the individual human body frame information to be processed, the method further includes the steps of:

and when the face advancing direction information and the human body advancing speed information accord with preset conditions, obtaining a door opening intention result corresponding to the target face identification.

Specifically, when the face advancing direction information and the human body advancing speed information accord with preset conditions, the server obtains a door opening intention result corresponding to the target face identification.

In the above embodiment, the face traveling direction information corresponding to the face identifier to be processed is determined by the area of each face frame to be processed. And then, whether a door opening intention result exists is determined according to the face advancing direction information and the human body advancing speed, namely, whether the door opening intention exists in the human body is judged through the face, so that the door opening intention detection method is more convenient and accurate than the method for judging by using the human body, and the accuracy of door opening intention detection is improved.

In one embodiment, as shown in fig. 4, the step S208 of determining the human body travel speed information corresponding to the human body identifier to be processed according to the individual human body frame information to be processed and the time interval includes the steps of:

s402, calculating each human body coordinate according to the position of the human body frame in each piece of to-be-processed human body frame information, and determining each speed according to each human body coordinate and the time interval.

The human body coordinates refer to coordinates of the center point of the human body frame.

Specifically, the server obtains vertex coordinates of each human body frame according to the position of the human body frame in each piece of to-be-processed human body frame information, for example, four vertex coordinates of the human body frame can be obtained by calculation with a vertex at a lower left corner of a video frame picture as an origin. And then calculating the coordinates of the center point of each human body frame according to the coordinates of the four top points of each human body frame to obtain the coordinates of each human body. And calculating to obtain each speed by using a speed calculation formula according to each human body coordinate and the time interval. For example, if there are five video frames in the video frame sequence, the five body coordinates corresponding to the five body frames corresponding to the five video frames are calculated to be (-10,0), (-7,0), (-4,0), (-2,0), and (-1,0), and the time interval is 40 ms. Then v1 (-7+ 10)/0.04-75 is calculated; v2 (-4+ 7)/0.04-75; v3 (-2+ 4)/0.04-5; v4 (-1+ 2)/0.04-25.

S404, determining the acceleration according to each speed, and obtaining the human body advancing speed information corresponding to the human body identifier to be processed according to each speed and the acceleration.

Specifically, the server determines the acceleration corresponding to the human body according to each speed and time interval, the speeds and the accelerations are used as the human body advancing speed information corresponding to the human body identifier to be processed, the human body is abstracted into coordinate points, the human body advancing speed is calculated by using the coordinate points of each frame, and the accuracy of obtaining the advancing speed is improved. For example, the acceleration corresponding to the human body may be calculated according to the initial velocity, the final velocity, and the total time interval. In a specific embodiment, the acceleration a of the human body is calculated to be (2.5-7.5)/0.16-31.25 according to the above v 1-7.5 and v 4-2.5 for a total time interval of 40-4-160. That is, the human body is doing deceleration movement, as shown in fig. 5, the human body is moving forward and doing deceleration movement schematically, wherein, the human body acceleration calculated by obtaining five human body coordinate points according to the human body detection frames of five video frames is-31.25, and then the human body is doing deceleration movement. In another specific embodiment, as shown in fig. 6, if the human acceleration calculated from five human coordinate points obtained from the human detection frames of five video frames is positive, the human body performs acceleration motion of forward travel. In another specific embodiment, as shown in fig. 7, if the human acceleration calculated from five human coordinate points obtained from the human detection frames of five video frames is zero, the human body is moving at a constant speed in the forward direction.

In one embodiment, after step S204, that is, after determining the human body travel speed information corresponding to the human body identifier to be processed according to the each human body frame information to be processed and the time interval at the time interval of acquiring the monitoring video frame, the method further includes the steps of:

and when the human body advancing speed in the human body advancing speed information is a target value, obtaining a door opening intention result corresponding to the human body identifier to be processed.

Specifically, the target value means that the human body traveling speed is 0, that is, the human body is in a stationary state. When a person suddenly appears in front of the monitoring device and stands still, the person is indicated to have an intention to open the door. At the moment, the server determines that the human body advancing speed in the human body advancing speed information is a target value, and directly obtains a door opening intention result corresponding to the human body identifier to be processed. Therefore, the advancing direction of the human body does not need to be determined, and the efficiency of obtaining the door opening intention result is improved.

In one embodiment, after step S210, that is, after the face image to be processed is matched with the preset face library, the method further includes the steps of:

and when the matching of the face image to be processed and the preset face library fails, generating alarm information, and sending the alarm information to the alarm equipment so that the alarm equipment displays the alarm information.

Specifically, the server matches the face image to be processed with a preset face library, and when the matching of the face image to be processed and the preset face library fails, the server generates alarm information which is used for indicating that a person without door opening authority wants to enter a door. And sending alarm information to alarm equipment, and displaying the alarm information after the alarm equipment receives the alarm information to remind a manager that the manager wants to enter but does not have permission. The alarm device can play the alarm information through voice.

In one embodiment, when the face image to be processed fails to match with the preset face library. The server matches the face image to be processed with the blacklist face database, when the face image to be processed is successfully matched with the blacklist face database, the server generates alarm information, the alarm information comprises blacklist face identity information and the like, the alarm information is sent to alarm equipment, the alarm equipment displays the alarm information, and management personnel can use the alarm information conveniently.

In one embodiment, after step S208, i.e. after determining the human body traveling direction information corresponding to the target human body identifier according to the respective target human body frame information, the method further includes the steps of:

and when the human body advancing direction information and the human body advancing speed information do not meet the preset conditions, obtaining a result without door opening intention, and obtaining door opening forbidding information according to the result without door opening intention.

The absence of the door opening intention result means that the intention detection result indicates that the door opening intention does not exist. The door opening prohibition information is information that the door opening operation of the access control device is not required.

Specifically, when the server judges that the human body advancing direction information and the human body advancing speed information do not meet the preset conditions, the server obtains the result that the door opening intention does not exist, and obtains the door opening prohibition information according to the result that the door opening intention does not exist, namely when the fact that the human body does not have the door opening intention is detected, the server stops subsequent processing for recognizing the human face corresponding to the human body, and the processing efficiency of the server is improved.

In one embodiment, after step S212, that is, after obtaining the door opening instruction according to the face recognition result and the door opening intention result, and sending the door opening instruction to the door control device, so that the door control device performs a door opening operation according to the door opening instruction, the method further includes the steps of:

and acquiring a face identity from the face recognition passing result, generating a door opening record according to the face identity, and storing the door opening record.

Specifically, the server acquires the face identity from the face recognition passing result, and generates a door opening record according to the face identity, wherein the door opening record records that the face identity enters the access control equipment at the current system time. The server stores the door opening record in the record database, so that subsequent checking and processing are facilitated.

In a specific embodiment, as shown in fig. 8, a timing chart of the access control method is shown. The method comprises the steps that a monitoring device sends an original video stream to a stream taking module of a server, the stream taking module of the server converts the original video stream into a monitoring video frame sequence, then the monitoring video frame sequence is sent to a face detection module of the server to carry out face detection to obtain face information, the monitoring video frame sequence is sent to a human body detection module of the server to carry out human body detection to obtain human body information, the human body detection module of the server sends the human body information to an intention recognition module to recognize the human body information to obtain human body advancing direction information and human body advancing speed information, when the human body advancing direction information and the human body advancing speed information meet preset conditions, door opening intention is obtained, and the human body detection module of the server sends a door opening intention result to an entrance guard control module. At the moment, when the intention identification module obtains the intention of opening the door, the face detection module of the server sends the face information to the face identification module for face identification. When the face recognition passes, the face recognition module of the server sends the face recognition passing result to the access control module, the access control module of the server generates a door opening instruction according to the face recognition passing result and the door opening intention result, the door opening instruction is sent to the access control equipment, and the access control equipment receives the door opening instruction to perform door opening operation.

In a specific embodiment, as shown in fig. 9, a flow chart of the access control method is shown. The method specifically comprises the following steps: the monitoring camera sends the original video stream to a camera stream acquisition module in the server service module, and the camera stream acquisition module converts the original video stream into a video frame sequence. And sending the video frame sequence to an algorithm module for face detection, human body detection, intention detection and face identification. And respectively carrying out face detection on the video frame sequence in the algorithm module through a face detection module to obtain a face detection result. The method comprises the steps of carrying out human body detection on a video frame sequence through a human body detection module to obtain a human body detection result, sending the human body detection result to an intention detection module for intention detection to obtain an intention detection result, sending a face detection result to a face recognition module for face recognition when the intention detection result is a door opening intention result, and judging whether the door is opened or not. When the face recognition passes, the face recognition passing result and the door opening intention result are sent to the post-processing module, the door opening module in the post-processing module sends a door opening instruction to the access control equipment through the GPIO information according to the face recognition passing result and the door opening intention result, and the access control equipment performs door opening operation. Meanwhile, the door opening module can also generate door opening prompt information and send the door opening prompt information to the information display equipment, and the information display shows the door opening prompt information. As shown in fig. 10, a schematic diagram of successful door opening when a face of a person is brushed successfully is shown, wherein the information display device displays door opening prompt information, and the person can successfully open a door to enter. When the face recognition fails, the face recognition failing result is sent to the door opening module, the door opening module receives the face recognition failing result and generates the information display device without the access permission, the information display device displays the prompting information without the access permission, and as shown in fig. 11, when a person fails to swipe the face, the schematic diagram of the door opening failure is shown, wherein the information display device displays the prompting information without the access permission, and the person cannot enter the information display device.

It should be understood that although the steps in the flowcharts of fig. 2-4, 9 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-4, 9 may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least some of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 12, there is provided an access control apparatus 1200 including: an information determination module 1202, a speed determination module 1204, an intent derivation module 1206, a face recognition module 1208, and an access control module 1210. Wherein:

the information determining module 1202 is configured to acquire a monitoring video frame sequence, and determine a to-be-processed face image and information of each to-be-processed human body frame corresponding to the to-be-processed face identifier according to the monitoring video frame sequence;

a speed determining module 1204, configured to obtain a time interval of the monitoring video frame, and determine human body travel speed information corresponding to the human body identifier to be processed according to each piece of human body frame information to be processed and the time interval;

an intention obtaining module 1206, configured to determine, according to the to-be-processed human body frame information, human body advancing direction information corresponding to the to-be-processed human body identifier, and obtain a door opening intention result corresponding to the to-be-processed human body identifier when the human body advancing direction information and the human body advancing speed information meet a preset condition;

the face recognition module 1208 is used for matching the face image to be processed with a preset face library when the result of the intention to open the door is obtained, and obtaining a face recognition passing result when the matching is successful;

and the access control module 1210 is configured to obtain a door opening instruction according to the face recognition pass result and the door opening intention result, and send the door opening instruction to the access control device, so that the access control device executes a door opening operation according to the door opening instruction.

In one embodiment, the information determination module 1202 includes:

the face recognition module is used for acquiring a monitoring video frame sequence, carrying out face detection on the monitoring video frame sequence to obtain each face identifier and each face image, acquiring a face identifier to be processed from each face identifier, and determining a face image to be processed corresponding to the face identifier to be processed from each face image;

and the human body detection module is used for carrying out human body detection on the monitoring video frame sequence to obtain each human body identifier and each human body frame information, determining the human body identifier to be processed from each human body identifier according to the human face identifier to be processed, and determining each human body frame information to be processed corresponding to the human body identifier to be processed from each human body frame information.

In one embodiment, a face detection module comprises:

the face detection unit is used for inputting the monitoring video frame sequence into a trained face detection model to carry out face detection so as to obtain each face identification, each face frame image and each corresponding face frame area;

the target face determining unit is used for determining the area of each face frame to be processed corresponding to the face image to be processed from the area of each face frame, and determining the area of the target face frame from the area of each face frame to be processed;

the face recognition module 1208 includes:

and the face recognition passing unit is used for inputting the face image to be processed into the trained face recognition model for face feature recognition when the area of the target face frame exceeds a preset threshold value to obtain a face feature recognition result, matching the face feature recognition result with face features in a preset face library, and obtaining a face recognition passing result when the matching is successful.

In one embodiment, the target face determination unit further includes:

the face direction determining unit is used for determining face advancing direction information corresponding to the face identification to be processed according to the area of each face frame to be processed;

then the intent is to get block 1206, further including:

and the face judgment unit is used for obtaining a door opening intention result corresponding to the target face identification when the face advancing direction information and the human body advancing speed information accord with preset conditions.

In one embodiment, the speed determination module 1204 includes:

the coordinate calculation unit is used for calculating each human body coordinate according to the position of the human body frame in each piece of human body frame information to be processed and determining each speed according to each human body coordinate and the time interval;

and the acceleration calculation unit is used for determining acceleration according to each speed and obtaining human body advancing speed information corresponding to the human body identifier to be processed according to each speed and acceleration.

In one embodiment, the intention obtaining module 1206 is further configured to obtain a door opening intention result corresponding to the human body identifier to be processed when the human body traveling speed in the human body traveling speed information is a target value.

In one embodiment, after the face recognition module 1208, the method further includes:

and the alarm module is used for generating alarm information and sending the alarm information to the alarm equipment when the matching of the face image to be processed and the preset face library fails so that the alarm equipment can display the alarm information.

In one embodiment, the intent derivation module 1206 further comprises:

and the door opening forbidding unit is used for obtaining the result of no door opening intention when the human body advancing direction information and the human body advancing speed information do not accord with the preset conditions, and obtaining the door opening forbidding information according to the result of no door opening intention.

In one embodiment, the door access control device 1200 further includes:

and the recording module is used for acquiring the face identity from the face recognition passing result, generating a door opening record according to the face identity and storing the door opening record.

FIG. 13 is a diagram illustrating an internal structure of a computer device in one embodiment. The computer device may specifically be the server 104 in fig. 1. As shown in fig. 13, the computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing identification list information, address information and the like of the access control device. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement an access control method.

Those skilled in the art will appreciate that the architecture shown in fig. 13 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, the access control device provided by the present application may be implemented in the form of a computer program, and the computer program may be run on a computer device as shown in fig. 13. The memory of the computer device may store various program modules constituting the access control apparatus, such as an information determination module 1202, a speed determination module 1204, an intention acquisition module 1206, a face recognition module 1208, and an access control module 1210 shown in fig. 12. The computer program constituted by the program modules causes the processor to execute the steps of the access control method according to the embodiments of the present application described in the present specification.

For example, the computer device shown in fig. 13 may execute step S202 by the information determination module 1202 in the access control apparatus shown in fig. 12. The speed determination module 1204 performs step S204. The intent block 1206 performs step S206. The face recognition module 1208 performs step S208. The door access control module 1210 performs step S210.

In one embodiment, a computer device is provided, which includes a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, causes the processor to execute the steps of the access control method. Here, the steps of the access control method may be the steps of the access control method in each of the above embodiments.

In one embodiment, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the computer program causes the processor to execute the steps of the access control method. Here, the steps of the access control method may be the steps of the access control method in each of the above embodiments.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by instructing the relevant hardware through a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. An access control method, comprising:

acquiring a monitoring video frame sequence, and determining a face image to be processed corresponding to a face identifier to be processed and information of each body frame to be processed according to the monitoring video frame sequence;

acquiring time intervals of monitoring video frames, and determining human body advancing speed information corresponding to the human body identification to be processed according to the human body frame information to be processed and the time intervals;

determining human body advancing direction information corresponding to the human body identification to be processed according to the information of each human body frame to be processed, and obtaining a door opening intention result corresponding to the human body identification to be processed when the human body advancing direction information and the human body advancing speed information meet preset conditions;

2. The method according to claim 1, wherein the determining the to-be-processed face image and the to-be-processed body frame information corresponding to the to-be-processed face identifier according to the sequence of monitoring video frames comprises:

acquiring a monitoring video frame sequence, carrying out face detection on the monitoring video frame sequence to obtain each face identification and each face image, acquiring a face identification to be processed from each face identification, and determining a face image to be processed corresponding to the face identification to be processed from each face image;

and performing human body detection on the monitoring video frame sequence to obtain each human body identifier and each human body frame information, determining the human body identifier to be processed from each human body identifier according to the human face identifier to be processed, and determining each human body frame information to be processed corresponding to the human body identifier to be processed from each human body frame information.

3. The method of claim 2, wherein the performing face detection on the sequence of monitored video frames to obtain respective face identifiers and respective face images comprises:

inputting the monitoring video frame sequence into a trained face detection model for face detection to obtain face identifications, face frame images and corresponding face frame areas;

determining the area of each face frame to be processed corresponding to the face image to be processed from the area of each face frame, and determining the area of a target face frame from the area of each face frame to be processed;

matching the face image to be processed with a preset face library, and obtaining a face recognition passing result when the matching is successful, wherein the face recognition passing result comprises the following steps:

when the area of the target face frame accords with a preset area condition, the face image to be processed is input into a trained face recognition model to be subjected to face feature recognition, a face feature recognition result is obtained, the face feature recognition result is matched with face features in a preset face library, when the matching is successful, a face recognition passing result is obtained, the area of the target face frame accords with the preset area condition and comprises at least one of the fact that the area of the target face frame exceeds a preset area threshold value and the area of the target face frame equals to a preset target area.

4. The method according to claim 3, further comprising, after determining each face frame area to be processed corresponding to the face image to be processed from the each face frame area:

determining face advancing direction information corresponding to the face identification to be processed according to the area of each face frame to be processed;

after the human body advancing direction information corresponding to the human body identifier to be processed is determined according to the information of each human body frame to be processed, the method further comprises the following steps:

5. The method according to claim 1, wherein the determining the human body travel speed information corresponding to the human body identifier to be processed according to the individual human body frame information to be processed and the time interval comprises:

calculating each human body coordinate according to the position of the human body frame in each piece of to-be-processed human body frame information, and determining each speed according to each human body coordinate and the time interval;

and determining acceleration according to each speed, and obtaining human body advancing speed information corresponding to the human body identifier to be processed according to each speed and the acceleration.

6. The method according to claim 1, wherein after the time interval of the acquisition of the monitoring video frame and the determination of the human body traveling speed information corresponding to the human body identifier to be processed according to the individual human body frame information to be processed and the time interval, the method further comprises:

7. The method according to claim 1, after the matching the face image to be processed with a preset face library, further comprising:

and when the matching between the face image to be processed and the preset face library fails, generating alarm information, and sending the alarm information to alarm equipment so that the alarm equipment displays the alarm information.

8. The method according to claim 1, after determining the human body traveling direction information corresponding to the target human body identifier according to the respective target human body frame information, further comprising:

and when the human body advancing direction information and the human body advancing speed information do not meet preset conditions, obtaining a result without door opening intention, and obtaining door opening prohibition information according to the result without door opening intention.

9. The method of claim 1, wherein after obtaining a door opening instruction according to the face recognition result and the door opening intention result and sending the door opening instruction to an access control device so that the access control device performs a door opening operation according to the door opening instruction, the method further comprises:

10. An access control apparatus, comprising:

the face recognition module is used for matching the face image to be processed with a preset face library when the result of the intention of opening the door is obtained, and obtaining a face recognition passing result when the matching is successful;

and the access control module is used for obtaining an opening instruction according to the face recognition passing result and the opening intention result and sending the opening instruction to the access control equipment so that the access control equipment executes opening operation according to the opening instruction.

11. A computer-readable storage medium, storing a computer program which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 9.

12. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 9.